Road conditions affect the safety of road users as potholes may cause traffic accidents. We design and implement a sensor-based application for road conditions detection. The sensor device includes a GPS NEO-6M module to get the pothole locations and an MPU-92/65 module as accelerometer and gyroscope sensor, which are attached to a NodeMCU ESP8266. We use REST API as the web service to connect the sensor device and users, who will access the information using the Android application on their smartphones. To cluster the acquired data, we use the k-means clustering algorithm. The algorithm clusters data into deep, medium and shallow potholes.